me2
2012-09-18 13:20:43 UTC
me2 [http://community.zenoss.org/people/me2] created the discussion
"looking for suggestions on eliminating a bad polling cycle skewing graphs"
To view the discussion, visit: http://community.zenoss.org/message/68641#68641
--------------------------------------------------------------
hello.
v3.2.1 on centos vm box.
Plenty of free memory
Very low cpu load on localhost
The destination being polled is in the middle of the night so load is low on that box also.
I have 4 data points being graphed on one report.
They are gauges and rps (requests per second).
The one I am having problems with is blocks per second. Normally this is low, between 1-30 depending on how busy an interface is.
On one specific oid, I get a poilling result that is way below the minimum number I have told it...0.
The issue I have is when this shows up it ruins the graph because it skews all the other number from other data points. Once I deleted the rrd the others come back of course and this one starts to rebuild. I have attached a graph showing this. Does anyone have ideas on how to eliminate this? The results can't be a negative number.
There is no consistency as to what device I am polling when this value is returned. I received it back 4-5 times out of 50 devices in a 24 hr period.
I am ruling out network latency because the other oids on the boxes come back fine.
thanks for your ideas.
Jay
the graph is attached
--------------------------------------------------------------
Reply to this message by replying to this email -or- go to the discussion on Zenoss Community
[http://community.zenoss.org/message/68641#68641]
Start a new discussion in zenoss-users by email
[discussions-community-forums-zenoss--***@community.zenoss.org] -or- at Zenoss Community
[http://community.zenoss.org/choose-container!input.jspa?contentType=1&containerType=14&container=2003]
"looking for suggestions on eliminating a bad polling cycle skewing graphs"
To view the discussion, visit: http://community.zenoss.org/message/68641#68641
--------------------------------------------------------------
hello.
v3.2.1 on centos vm box.
Plenty of free memory
Very low cpu load on localhost
The destination being polled is in the middle of the night so load is low on that box also.
I have 4 data points being graphed on one report.
They are gauges and rps (requests per second).
The one I am having problems with is blocks per second. Normally this is low, between 1-30 depending on how busy an interface is.
On one specific oid, I get a poilling result that is way below the minimum number I have told it...0.
The issue I have is when this shows up it ruins the graph because it skews all the other number from other data points. Once I deleted the rrd the others come back of course and this one starts to rebuild. I have attached a graph showing this. Does anyone have ideas on how to eliminate this? The results can't be a negative number.
There is no consistency as to what device I am polling when this value is returned. I received it back 4-5 times out of 50 devices in a 24 hr period.
I am ruling out network latency because the other oids on the boxes come back fine.
thanks for your ideas.
Jay
the graph is attached
--------------------------------------------------------------
Reply to this message by replying to this email -or- go to the discussion on Zenoss Community
[http://community.zenoss.org/message/68641#68641]
Start a new discussion in zenoss-users by email
[discussions-community-forums-zenoss--***@community.zenoss.org] -or- at Zenoss Community
[http://community.zenoss.org/choose-container!input.jspa?contentType=1&containerType=14&container=2003]